Low-Latency Decisioning Engineer

Mylitm; California (MO); On-site; USD 120,000 - 180,000; Full time

Job summary

Autonomous Decisioning Research & Engineering seeks an Automated Decisioning Engineer to design and build systems that transform raw inputs into defensible decisions within milliseconds. You will balance model outputs, deterministic rules, regulatory constraints, and human review queues into a coherent pipeline.

Responsibilities include architecting pipelines, building APIs with strict latency SLAs, and implementing audit trails and explanations.

Qualifications

• 5+ years backend or platform engineering, with at least 2 in ML-adjacent systems.
• Strong systems design skills — you have built low-latency, high-availability services.
• Working knowledge of rules engines (Drools, OpenL Tablets) or workflow orchestration (Temporal, Cadence, Airflow).
• Comfort with distributed systems concepts (idempotency, exactly-once, backpressure).
• Excellent written communication — your design docs make decisions easier.

Responsibilities

• Architect decisioning pipelines that combine model inference, rules engines, and human review.
• Build APIs and event-driven services that meet strict latency and reliability SLAs.
• Implement audit trails, decision explanations, and override mechanisms.
• Partner with research to translate model outputs into customer-impacting decisions.
• Drive design reviews, on-call quality, and post-mortems for decisioning incidents.

Skills

Backend engineering

ML-adjacent systems

Distributed systems

Rules engines / workflow orchestration

Technical communication

Tools

Drools

OpenL Tablets

Temporal

Cadence

Airflow

Job description

Autonomous Decisioning Research & Engineering seeks an Automated Decisioning Engineer to design and build systems that transform raw inputs into defensible decisions within milliseconds. You will balance model outputs, deterministic rules, regulatory constraints, and human review queues into a coherent pipeline.

Responsibilities include architecting pipelines, building APIs with strict latency SLAs, and implementing audit trails and explanations.